of the disincentives which appeared to be the most real or were mentioned with the greatest frequency. Perhaps the primary disincentive is the fear that use of unit costs would tend to produce tighter budgets because it would make it easier for reviewers to understand the programs and to reduce budgets to minimum levels. Tighter budgets are not only more difficult for a manager to operate under but also produce other disincentives. In this respect, there is a prevailing belief that in Government the rewards in terms of prestige, promotions, and the ability to hire better people and to carry out more effective programs, are awarded to those who can justify the largest budgets. Officials are reluctant to expose their budgets to reductions by adopting unit costs without compensating incentives. A second factor of equal or greater significance is the belief among agencies that use of unit costs in the budget process would bring about lower quality program output and loss in program effectiveness. This fear seems to stem from the feeling that use of unit costs would cause reviewing authorities to become overly concerned with efficiency to the detriment or disregard of program effectiveness, and that the unit costs would be used to effect unjustified budget reductions. The officials also fear that the unit cost measures may not be realistic and, thus, could be improperly used to justify budget cuts which would reduce program effectiveness. A third and related factor is the belief that the use of unit costs in its budget would build rigidity into agency programs and its management would make it more difficult for the agency to respond to unforeseen changes in situations and to congressional or public wishes for program changes. A fourth disincentive concerns the question of cost versus benefits. The establishment and maintenance of a cost accounting system needed to develop good unit costs is not without a cost of its own. Many agencies feel that it is not worth the added benefits or information that unit costs would provide. In particular, many agencies with well developed work measurement systems believe that unit cost measures would not provide them with sufficient additional information to warrant the added costs. Another related factor is the rather generally accepted proposition that there is no critical need at lower management levels in the organization for unit costs to manage low volume operations. Evidence indicates that where such information has been needed it has been developed. con One of the most frequently mentioned difficulties detering wider use of unit costs is that some programs are sidered very difficult, if not impossible, to measure in terms of unit costs and in particular it is questioned whether the results would be meaningful. Examples are research programs and some types of grant-in-aid programs. There seems to be some built-in reluctance of people to attempt to measure programs for which they have responsibility. This perhaps relates to the fact that people tend not to like to be measured. In a special study performed for the joint project it was found that some of these fears which act as disincentives to the use of unit costs were unfounded. In very brief terms one conclusion of the study was that users of unit costs in the preparation and support of their budget requests certainly fare no worse in budget reviews than non-users, and there is considerable evidence that they fare better. Extending the Use of Unit Costs With the available evidence on the benefits of unit costs as a basis for sound decisionmaking, there seems little doubt that extending their use would be a productivity enhancing endeavor. The problem of getting more extended use of unit costs seems to be primarily one of providing sufficient incentives to managers to use this tool to offset their apprehensions about making their operations more visible to higher management levels and budget examiners. The type of incentive the team deems most practical is that of greater managerial freedom. As indicated previously, there are limitations on managers which prevent them from making certain types of decisions notwithstanding how much more productive an alternative arrangement might be. Such limitations are personnel ceilings which prevent managers from hiring additional personnel in cases where such hiring would improve productivity. Other limitations are appropriation limitations which prevent money appropriated for one purpose from being used for another no matter how extreme the exigency or potential saving. Another matter which the joint team recognized as needed to provide managers with incentive for effective use of unit costs was the relating of budgetary costs with actual costs. As the situation now generally exists in Government agencies, budgeting and accounting for costs are not correlated effectively. If unit costs or similar data are used to justify budgetary presentations there is no follow-through to the actual costs incurred after the budget has been approved. Once the budget has been approved, the details of the data used in justifying it is forgotten and is not used as a control over operations in the budget execution phase. The joint team believes that this is largely because the agencies are not held accountable for how well they follow-through on budget objectives once the budget has been approved. The team believes that managers might be motivated to further use of unit costs if they could manage on the basis of cost without other restraints. The team also believes that extended use of unit costs could be attained by greater use of such data in the budgetary process. Accordingly, we are making the following recommendations. Recommendations We recommend that: 1. The OMB establish a pilot project in which a selected volunteer agency is permitted to have maximum potential flexibility in managing on the basis of unit costs and other valid measurement data and be relieved of personnel ceilings, grade controls and other arbitrary restraints. The objective is to es tablish whether this is not a superior means of controlling Federal managers while still permitting managers maximum flexibility to adjust their operations to maximize productivity. 2. The OMB enforce Circular A-11 requirements that unit costs be used in budget submissions to justify appropriation requests where such costs can be prepared from existing data. The objective is to promote the use of better measurement data in the process by which resources are made available to agencies. 3. A demonstration project be undertaken under the ENCOURAGING EFFECTIVENESS MEASUREMENT The purpose of this project was to determine the extent to which Federal agencies measure the impact or effect of their programs and to encourage more valid and useful means of measuring effectiveness. It was found in Phase I of this project that most efforts to evaluate programs or organizations have been concerned with measuring productivity in terms of input-output relationships. The outputs are usually expressed in units of goods or services immediately produced, such as number of claims processed, number of persons trained, etc. These measures assume that the activity is doing what it should be doing and provide no means for evaluating the contribution or success of the activity relative to its cost.1 Therefore, this project was added in Phase II to emphasize the importance of developing effective measures in conjunction with productivity measures. Most of the existing measurement systems deal with aspects of organizational efficiency, which are determined by comparing actual performance (unit costs, for example) with some standard. The major problem in progressing from efficiency to effectiveness measurement is that the latter involves establishment of complex external and internal causeeffect relationships with the external tending to be the more difficult to determine. For example, the mission of the National Highway Safety Bureau, DOT, is "to reduce the mounting number of deaths and injuries resulting from traffic accidents on the Nation's highways." Establishing a direct relationship between the outputs of the Bureau's three programs and the incidence of traffic deaths and injuries appears feasible, but not easy. Effectiveness measurement provides the means of determining whether the agency is proceeding toward the objectives and of establishing a relationship between management actions and mission accomplishment. Both efficiency and effectiveness measurements are thus essential tools of managers in assessing true productivity, the former determining the cost of producing the agency's outputs and the latter the value of the agency's outputs to the recipient of its goods and services. 1 For a complete discussion of the interrelationship between all measures see Appendix I. Measures of effectiveness will provide information on: --Whether programs are accomplishing their intended --Ways to improve the operation of the program. --Which programs should be abolished and which new programs should be undertaken. --How well programs are operating for which no final output can be readily defined. In developing recommendations for encouraging the utilization of effectiveness measurement, we have attempted to profit from the experiences of Programming Planning and Budgeting (PPBS). In summary, the implementation of a Government-wide PPB system in the mid-60's had the following deficiencies: --Rigidly forced from the top-down, on a Governmentwide basis, it was adopted rather than adapted. --Not well understood, supported, or used by top management. --Perceptions of PPB's purposes not clear at all levels (OMB, Agency heads, managers). --Little impact on resource allocation decisions, and not related to any system of rewards or penalties. --Not integrated with the decisionmaking process Unfortunately, PPBS was often viewed as a single, welldefined system based on sophisticated techniques of analysis and intended to produce an optimal plan of Government activity. Since the introduction of PPBS, there have been several attempts to develop effectiveness measurement for program evaluation and analysis. For example, in President Nixon's May 1970 memorandum to agency heads, he asked for program evaluation involving three steps: --First, a critical examination of the objectives of the program. --Second, an analysis of the effectiveness of the program. Does the program adequately serve its target |